Human pluripotent stem (hPS) cells have great potential as a renewable source of cells for drug discovery, disease modeling and for transplantation to replace cells lost to degenerative diseases or injury. A critical barrier to the successful application of hPS cell research is the lack of methods for producing pure populations of hPS cell derivatives. Contaminating cells are both an issue for drug discovery where pure populations are needed for assay reproducibility and for transplantation where they present a safety issue because of their potential for tumor formation or differentiation to inappropriate cell types. Specifically, we propose here to address the problem of residual contaminating undifferentiated hPS cells that have the potential to form teratomas. We propose to develop hPS cell targeted toxins (hPS-CTTs) that kill residual hPS cells with minimal damage or alteration of the differentiated cell population. To achieve selective cell targeted toxicity, w will combine an internalizing hPS selective targeting agent with the plant toxin, saporin, which is only toxic upon cellular internalization. This approach is much simpler than previously proposed genetic modification and physical separation methods that require cell manipulations that could alter or stress the desired cell population. It is also much more flexible than the previously proposed cytotoxic antibody because any internalizing antibody or peptide can be adapted for targeted cell killing. Moreover, the hPS-CTT is designed to be removed with the dead cells thus reducing its potential impact on the surviving cells. To demonstrate feasibility in phase I, we wil screen antibodies and peptides for selective hPS cell targeting. The cell targeting agents (CTAs) will be indirectly conjugated to a toxin, and tested for their ability to remove undifferentiated hS cells in a model cell system containing admixtures of differentiated hPS cells spiked with undifferentiated hPS cells. We will measure hPS cell content before and after treatment by flow cytometry, the PluriTest, and quantitative RT-PCR. The ability to the optimal hPS-CTT to remove hPS cells will be tested using an in vivo teratoma assay. The persistence of the hPS-CTT in the treated population will be measured by western blot analysis and a functional assay to detect residual toxicity. A 2 component system that combines various targeting agents with secondary toxin conjugates will be used in phase I to show feasibility. In phase II, we will develop cell targeting agents directly conjugated to the toxin and further optimize protocols for hPS cell removal.